Skip to main content
2025 SCMR Conference Main banner
Final Program


Schedule at a Glance
Back
Video On-Demand Video

ISMRM - SCMR Workshop

ISMRM - SCMR Workshop: Power Pitch Oral Session

Power Pitch 3: Flip-back Motion Compensated Spin Echo Imaging for Reduced Field of View Multi-slice In Vivo Diffusion Tensor Cardiovascular Magnetic Resonance

Wednesday, January 29, 2025
4:06 PM – 4:09 PM East Coast USA Time
Location: Blue Room
Background:

Diffusion Tensor Cardiovascular Magnetic Resonance (DT-CMR) is an effective technique for gaining insights into in vivo myocardial microstructure [1-2]. Shorter Echo Planar Imaging (EPI) readouts are desirable to reduce artefacts. This can be achieved with a reduced Phase Encoding (PE) Field of View (FOV), by making the 90° RF pulse slice selective in the PE direction. SNR efficiency can be enhanced by interleaved multi-slice imaging, but it is incompatible with slice selective 90° RF pulse along PE due to saturation. One solution is to use 2D RF pulses [3], however, they can increase TE and unwanted side-lobe excitation outside the desired FOV, leading to wrap. Here we demonstrate an alternative method which limits the PE FOV using a slice selective gradient on the 180° refocusing pulses and adds an additional 180° pulse after the readout [4-5], which restores the inverted magnetisation outside the imaged slice. This study is the first to investigate this technique in DT-CMR for boosted SNR and reduced FOV multi-slice acquisition.

Methods:

Four different approaches to reduce the PE FOV were compared (Fig. 1):


  1. 90° RF pulse (PE) and 180° (Slice Selection (SS)) [6].

  2. 90° RF pulse (SS) and 180° (PE).

  3. The proposed flip-back pulse sequence: 90° RF pulse (SS), then 180° (PE) with an additional flip-back 180° after the EPI readout. Outside the imaged slice, the first 180˚ pulse inverts the magnetisation and the second pulse flips it back to its original state.

  4. 90° 2D RF pulse, to excite on both the SS and PE directions (ZoomIt), then 180° (SS).


DT-CMR images using multi-slice second order Motion Compensated Spin Echo (MC-SE) EPI were acquired from 6 controls (3 for method optimisation, 3 for evaluation) in systole at 3T (Siemens Cima.X) with resolution 2.8x2.8x8mm3, Field of View (FOV) 135x360mm2, b0 data and six encoding directions at b equals 450 s/mm2 with 8 averages. Two slices were acquired simultaneously. This is time-efficient, as 14 images can be acquired in one 18 cardiac cycle breath hold. All sequences with TR = 2 RR-intervals, sequences 1-3 with TE = 50ms, sequence 4 with TE = 67ms. The proposed sequence was validated using an agar phantom, with T1 = 1969ms, T2 = 720ms.

Results:

Examples of DT-CMR images for the four two-slice sequences are shown in Fig. 2, and SNR values from the 6 subjects and mean DT-CMR values from 3 subjects are presented in Fig. 3. Mean SNR values for phantom are 3.65, 3.01, 4.00, 3.97 for sequences 1-4. Similarly, the mean SNR values for in vivo data are 4.57, 4.36, 6.32, 5.57 respectively, with an increase of 38% from PE90 sequence to the proposed sequence. The mean DT-CMR values are similar across 2DRF and proposed sequence. 

Conclusion:

The proposed flip-back multi-slice pulse sequence for in vivo MC-SE DT-CMR, reduces the PE FOV using a slice selective 180° refocusing pulse and uses a second RF pulse after the readout to “flip-back” the inverted magnetisation. This can allow for boosted SNR, and efficient acquisition of multi-slice data. In future studies, the slice profiles of the 180° pulses may be further optimised to deliver more homogeneous SNR improvement.

Our approach has potential future clinical utility in enhancing image quality and increasing acquisition efficiency of DT-CMR in patient cohorts.